A Quadric Error Metrics Based Feature Reserving Simplification Method
Zhe Guo · Chinese Journal of Stereology and Image Analysis · 2009
Since 3D face models usually have very high resolution and uneven feature distribution,mesh should be simplified before processing to reduce redundancy information in the model and improve computing efficiency.However,current mesh simplification algorithms are not satisfying in feature preservation.In order to reserve information among feature rich area,the constant mean curvature of each vertex was used to compute its collapse cost along with geometrical error measurement.At the same time,this improved method added examination and collapse steps to solve the overlapping faces problem in Quadric Error Metrics simplification.Experiments demonstrated that this method can reserve detailed information as well as reliably producing high-quality simplified mesh.